• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸°úÇÐȸ ³í¹®Áö > Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Current Result Document : 28 / 29

ÇѱÛÁ¦¸ñ(Korean Title) Ç׸ñ Àα⵵ ÆíÇâ °üÁ¡¿¡¼­ÀÇ ÀâÀ½Á¦°Å ¿ÀÅäÀÎÄÚ´õÀÇ È¿°ú
¿µ¹®Á¦¸ñ(English Title) Effect of Denoising Autoencoder in the view of Item Popularity Bias
ÀúÀÚ(Author) ¹ÚÁø¿µ   ¿ÀÇý¿¬   JinYeong Bak   Alice Oh   ±èÁøÈ«   ÀÌÀç¿õ   ÀÌÁ¾¿í   Jinhong Kim   Jae-woong Lee   Jongwuk Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 48 NO. 05 PP. 0575 ~ 0583 (2021. 05)
Çѱ۳»¿ë
(Korean Abstract)
ÀâÀ½Á¦°Å ¿ÀÅäÀÎÄÚ´õ´Â Ãßõ ½Ã½ºÅÛ¿¡¼­ ÃÖ±Ù ÈçÈ÷ »ç¿ëµÇ°í ÀÖ´Â ¸ðµ¨ÀÌ´Ù. ÀÌ ¸ðµ¨Àº ÀԷ¿¡ ÀâÀ½À» ÁÖ¾î ÇнÀ½ÃÅ°´Â ¿ÀÅäÀÎÄÚ´õÀÇ ½Å°æ¸Á ±â¹Ý Ãßõ ¸ðµ¨·Î ¿ÀÅäÀÎÄÚ´õ¿¡ ºñÇØ ³ôÀº Á¤È®µµ¸¦ º¸ÀδÙ. º» ³í¹®¿¡¼­´Â ÀâÀ½Á¦°Å ¿ÀÅäÀÎÄÚ´õÀÇ ÇнÀ °úÁ¤À» ÀÌÇØÇϱâ À§Çؼ­, Ç׸ñÀÇ Àα⵵ ÆíÇâ °üÁ¡¿¡¼­ ÀâÀ½ÀÇ È¿°ú¸¦ ºÐ¼®ÇÑ´Ù. ºÐ¼®À» À§ÇØ ¿ì¸®´Â ´ÙÀ½ÀÇ µÎ °¡Áö ¹æ¹ýÀ¸·Î ½ÇÇèÀ» ¼³°èÇÑ´Ù. ¿ì¼±, ¿ÀÅäÀÎÄÚ´õ¿¡ ÀâÀ½À» ÁÖ´Â ¹æ¹ýÀ¸·Î ÇнÀµÈ Ç׸ñ º¤ÅÍÀÇ L2 ³ë¸§(L2-norm)ÀÇ º¯È­¸¦ °üÂûÇÑ´Ù. ´ÙÀ½À¸·Î´Â, Ç׸ñÀÇ Àα⵵¿¡ ÀÇÇØ ÀÏÂ÷ÀûÀ¸·Î ÃßÃâµÈ Ç׸ñ¿¡¸¸ ÀâÀ½À» ÁÖ´Â ¹æ¹ýÀ» ÅëÇØ, ÀâÀ½Á¦°Å ¿ÀÅäÀÎÄÚ´õÀÇ ¼º´É Çâ»ó È¿°ú¿Í Ç׸ñÀÇ Àα⵵°£ °ü·Ã¼ºÀ» ºÐ¼®ÇÑ´Ù. ½ÇÇè°á°ú¸¦ ÅëÇØ Àα⵵¿¡ ÀÇÇØ »ý±ä Ç׸ñ º¤ÅÍ ³ë¸§ÀÇ ºÐ»êÀÇ Å©±â°¡ ÀâÀ½¿¡ ÀÇÇØ ÁÙ¾îµå´Â °ÍÀ» È®ÀÎÇÏ¿´À¸¸ç, ¶ÇÇÑ Àα⵵°¡ ³ôÀº Ç׸ñ¿¡ ÀâÀ½À» ÁÙ ¶§ Á¤È®µµ Çâ»ó¿¡ µµ¿òÀÌ µÇ´Â °ÍÀ» È®ÀÎÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Denoising autoencoder (DAE) is commonly used in recent recommendation systems. It is a type of Autoencoder that trains by giving noise to the input and has shown improved performance compared to autoencoder. In this paper, we analyze the effect of noise in terms of item popularity to interpret the training of DAE. For analysis, we design the experiment in the following two ways. First, we observe the changes of the learned item vector¡¯s L2-norm by giving noise to the autoencoder. Second, by giving noise only to presampled items by popularity, we anlayze whether the improved performance of the DAE is related to item popularity. Results of the experiment showed that the variance of the item vector norm caused by popularity was reduced by noise, and that the accuracy increased when noise was given to the popular items.
Å°¿öµå(Keyword) ºÐ·ù   ¸®´õ½Ê   RNN   BERT   Á¶¼±¿ÕÁ¶½Ç·Ï   classification   leadership   annals of the Joseon dynasty   ÀâÀ½Á¦°Å ¿ÀÅäÀÎÄÚ´õ   Çù¾÷ ÇÊÅ͸µ   »óÀ§-N Ãßõ   Ç׸ñ Àα⵵   denoising autoencoder   collaborative filtering   top-N recommendation   item popularity  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå